jialicheng/unlearn_cifar100_resnet-50_random_label_6_87

87

This model is a fine-tuned version of microsoft/resnet-50 on the cifar100 dataset. It achieves the following results on the evaluation set:

• Loss: 0.6196
• Accuracy: 0.8408
• Dt Accuracy: 0.8408
• Df Accuracy: 0.7967
• Unlearn Overall Accuracy: 0.3512
• Unlearn Time: None

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.0002
• train_batch_size: 128
• eval_batch_size: 256
• seed: 87
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Overall Accuracy	Unlearn Overall Accuracy	Time
No log	1.0	391	0.6446	0.87	0.2451	0.2451	None
1.4361	2.0	782	0.6270	0.8753	0.2371	0.2371	None
1.2783	3.0	1173	0.6289	0.8753	0.2371	0.2371	None
1.1975	4.0	1564	0.6368	0.8767	0.2351	0.2351	None
1.1975	5.0	1955	0.6268	0.8817	0.2274	0.2274	None
1.1314	6.0	2346	0.6284	0.8697	0.2461	0.2461	None
1.0842	7.0	2737	0.6163	0.8713	0.2435	0.2435	None
1.0419	8.0	3128	0.6246	0.8667	0.2507	0.2507	None
1.0013	9.0	3519	0.6237	0.8643	0.2542	0.2542	None
1.0013	10.0	3910	0.6186	0.8587	0.2628	0.2628	None
0.9664	11.0	4301	0.6181	0.8487	0.2777	0.2777	None
0.9455	12.0	4692	0.6247	0.8347	0.2983	0.2983	None
0.9189	13.0	5083	0.6186	0.831	0.3035	0.3035	None
0.9189	14.0	5474	0.6141	0.823	0.3147	0.3147	None
0.8961	15.0	5865	0.6180	0.812	0.3301	0.3301	None
0.8792	16.0	6256	0.6159	0.8067	0.3375	0.3375	None
0.8627	17.0	6647	0.6214	0.7933	0.3557	0.3557	None
0.8507	18.0	7038	0.6187	0.7963	0.3517	0.3517	None
0.8507	19.0	7429	0.6166	0.802	0.3439	0.3439	None
0.8365	20.0	7820	0.6196	0.7967	0.3512	0.3512	None

Framework versions

• Transformers 4.39.3
• Pytorch 2.2.2+cu118
• Datasets 2.18.0
• Tokenizers 0.15.2